Recent Research progress of Cal&Val Comprehensive Site and payload performance Assessment

Transcription

1 Academy of Opto-Electronics, Chinese Academy of Sciences Recent Research progress of Cal&Val Comprehensive Site and payload performance Assessment Lingling Ma Mar IVOS 25 Infrared and Visible Optical Sensors Subgroup to WGCV, CEOS March 19 21, 2013 at ESA ESRIN 1

2 Outlines Brief Introduction to AOE Overview of ETA Current work and achievements in quality assurance group Preparation work of WGCV 36 2

3 Brief Introduction to AOE Academy of Opto Electronics (AOE), established in 2003, is a high tech research institution of the Chinese Academy of Sciences (CAS), which is dedicated to coordinate CAS s institutes to work together and carry out some prospective or omitted research on the the development of space and opto electronic science and technology. AOE, with its headquarter based in Beijing, involves 5 independent research institutes of CAS located in 4 other cities of China. Changchun Institute of Optics, Fine Mechanics Physics Academy of Opto-Electronics Xi an Institute of Optics & Precision Mechanics HJ 1 HSI/IRS Tiangong 1/HSI Shanghai Institute of Technical Physics Shanghai Institute of Optics & Fine Mechanics Chengdu Institute of Optics & Electronics FY 3 VIRR/MERSI/IRAS/ERM Charg e 1 CCD/Imaging spectrometer /Laser altimeter 3

4 Brief Introduction to AOE The research activities of AOE focus on optical engineering, aeronautical and aerospace engineering and the application technology. The application technology consists of opto electric payload imaging mechanism and method, the opto electric payload performance evaluation and data quality monitoring technology. The research activities in this field include prospective research and system solution design and implementation. The aeronautical and aerospace engineering consists of space systems engineering, satellite navigation and aerostat technology. The research activities in this field include the research of development strategy, the prospective research, the System solution design and implementation, key technology and system integration. The optoelectronic engineering consists of computational optical imaging technology, projection optical system technology and large complex laser technology. The research activities in this field include prospective research, system solution design and implementation, key technology and system integration. 4

5 Overview of ETA The Earth Observation Technology Application Department (ETA) was established in 2005, responsible for the research area of The application technology in AOE. Engineering development of the data processing, operational management and emergency response system. Calibration and validation test sites construction, the methodology research on Cal&Val and payload performance analysis. Mechanism and application research on quantitative remote sensing, including hyperspectral, thermal infrared, LiDAR and SAR. 5

6 Current work and achievement in quality assurance group Comprehensive Cal&Val test sites Two comprehensive Cal&Val test sites were preliminarily established, which have different climate characteristics, various land cover types and topographic features. Now they are being continuously established so as to support multi grade validation of airborne and spaceborne sensors for stable and long term operation North China site South China site 6

7 Current work and achievement in quality assurance group Comprehensive Cal&Val test sites System components Auxiliary support systems 1 UAV Flying platform hyperspectral camera large field multispectral camera standard RS sensors 3 4 (a) artificial targets Standard targets (b) natural ground scenes 5 radiometric, geometric and spectral performance Data Processing and Analyzing System 7 Scientific Analysis Reports of Payload Performance

8 Atmospheric and field measurements Current work and achievements in quality assurance group Spectral reflectance of targets were measured. The aerosol optical thickness data and meteorological profile (including atmospheric temperature, atmospheric pressure, humidity, wind speed, wind direction, etc.) above the test site were synchronously collected :11:10 11:11:06 13:11:04 15:11:06 550nm aerosol optical thickness BRF 角度 ( ) MFT nm MFT nm MFT nm MFT nm MFT nm MFT nm MFT nm MFT nm MFT nm BRF properties of targets Spectral reflectance of gray-scale targets 实测靶标光谱反射率 Target reflectance by filed measurement Surface reflectance 60% 50% 40% 30% 20% 10% 0% Wave length(um) 50% 40% 30% 20% Radiosounding balloon :11:10 11:11:06 13:11:04 15:11:06 Moisture content Automatic sun tracking photometer, CE318 Automatic weather station 8

9 Inflight Calibration & Performance Assessment Absolute radiometric calibration Hyperspectral Gain Spectral band Multispectral Flow chart of optical sensor radiometric calibration Radiometric calibration coefficients have very good linearity and the correlation coefficient reaches above 99%. 9

10 Current work and achievements in quality assurance group Real central wavelength Central wavelength 中心波长 Spectral calibration for hyperspectral sensor with atmospheric absorption line 280~315 line Pixel line Pixel line Retrieval central wavelength based on flight data in Band 75 is O 2 absorption band. (laboratory measurement of Band 75 in Sep. 2010) (2) Shift of central wavelength in The shift of central wavelength is approximately 4~6nm compared to laboratory measurement. z 10

11 Current work and achievements in quality assurance group Spectral calibration for multispectral sensor with colored targets Difficulty: Solving of spectral response function faces ill condition matrix Solutions: The spectral reflectance of 15 multispectral targets were measured to add the number of equations; Piecewise fitting SRF according to laboratory measurements Laboratory Retrieval Response rate Response rate Spectral reflectance of colored targets Response rate wavelength (nm) Laboratory Retrieval Response rate wavelength (nm) wavelength (nm) wavelength (nm) 11

12 Current work and achievements in quality assurance group Payload performance analysis The payload performances were analyzed through several flight campaigns. SNR SNR Normalization of hyperspectral sensor in 2010/ MTF Band Number Year 2010 Year 2011 HSI MTF@Nyquist in 2010/2011 Band Number Dynamic range evaluation of hyperspectral sensor in 2010/2011 Dynamic Range (W/m2 um sr) High point in 2010 Low point in 2010 High point in 2011 Low point in Band Number Panchromatic image in 2011 Shift of central wavelength of hyperspectral sensor in (2) Absolute radiometric calibration Signal to Noise Ratio, SNR Dynamic range Response linear degree Radiometric resolution, NE ρ Ground resolution MTF Band registration precision central bandwidth; FWHM multispectral SRF 12 Reflectance/LAI retrieval z

13 Current work and achievements in quality assurance group Cal&Val method relative radiometric calibration Because of the variation in velocity height ratio, the images of hyperspectral imager between adjacent flight strips lack of comparability for different surfaces. A 3 D relative radiometric calibration method based on line frequency difference is proposed to solve this problem. 13

14 Current work and achievements in quality assurance group Cal&Val method Color calibration for hyperspectral image Why color calibration? True color composition image from arbitrarily selected hyperspectral channels located in the red, green and blue range will cause significant difference with comparison to the human visual. The degradation of the sensor performance will also have great impacts on the color composition. It s attractive to robustly and efficiently obtain the true color image from hyperspectral data. Spectral range: 400nm 1000nm Spectral resolution: 5nm 635 nm/545 nm/465nm 14

15 Current work and achievements in quality assurance group Cal&Val Color calibration for hyperspectral image True color image composition based on the spectral luminosity colorimetric theory The RGB values of the at sensor true color image is calculated with the hyperspectral images in the CIE XYZ space. The standard color space CIE XYZ by CIE TOA theoretical color acquisition based on radiative transfer simulation from field measurement With the in situ measurements of spectral reflectance, we can obtain the theoretical color according to the radiative transfer equation. In situ measured spectral reflectance of the targets 15

16 Current work and achievements in quality assurance group Cal&Val method Color calibration for hyperspectral image Color calibration model establishment A relationship between the theoretical colors and reconstructed colors is established and used to correct the image Theoretical color Reconstructed color with hyperspectral image MRGB RG ' ' B' True color correction coefficient matrix The corrected true colored image Composition Image before calibration 16

17 Current work and achievements in quality assurance group The extension of the Cal&Val test sites permanent targets In order to support in flight/orbit calibration and long term monitoring of payloads, we carried out wide investigations and now are establishing permanent targets for geometric and image quality test. Sjökulla ( FGI, Finnish Geodetic Institute) 17

18 Current work and achievements in quality assurance group The extension of the Cal&Val test sites Pre-flight calibration and performance test Considering the disassemble characteristic of airborne sensors, in situ high precision calibration and performance test system is being developed on the test site. Visible SWIR:MTF SNR resolution MRC uniformity MIR TIR:MDT MRDT MTF NETD SNR uniformity 基本控制点 标志点 Preflight geometric calibration 18

19 Current work and achievements in quality assurance group The extension of the Cal&Val test sites Spectrum extension to MIR and TIR Till now, the Cal&Val test sites can meet the requirements in radiometric, geometric and spectral Cal&Val of VIS, NIR, SWIR sensors The Cal&Val for MIR and TIR payload will be addressed next step. Ultra violet Visible Infrared NIR SWIR MIR TIR Microwave Now of the Cal&Val sites Future of the Cal&Val sites Next step for optical sensor (VIS, NIR, SWIR) Covering the visible, infrared and microwave band Supplement for SAR The Cal&Val for MIR and TIR calls for a simultaneous measurement of surface temperature distribution, emissivity spectrum of surface materials, atmospheric temperature and moisture profiles. 19

20 Current work and achievements in quality assurance group The extension of the Cal&Val test sites Spectrum extension to MIR and TIR Standard spectrometer imager It can be used in the field to simultaneously capture the temperature variations and emissivity and reflectivity spectrum for a large area rapidly. Standard spectrometer imager: Covering VIS, NIR, SWIR and TIR (to 12.5 μm) ; Spectral resolution: 1nm, 5nm, 30nm; Radiometric accuracy: 2%, 2%, 0.3K; GSD: 0.002m, 0.002m, Standard airborne spectrometer To assure the accuracy in the multilevel validation. Standard airborne spectrometer: Covering VIS, NIR, SWIR and TIR (to 12.5 μm) ; Spectral resolution: 5nm,10nm,60nm; Radiometric accuracy: 3%, 3%, 0.5K; The FOV is larger than 60 ; GSD: 0.5m, 0.5m, Anticipated traceable absolute radiometric calibration accuracy: 5.0%( μm); 1.0K(@300K)(8 12.5μm) 20

21 Current work and achievements in quality assurance group Hyperspectral thermal infrared data processing and retrieval Thermal infrared remotely sensed radiance, which is a measurement of emitted signals from both the surface and atmosphere, consists synthetic effects of land surface temperature, emissivity, atmospheric profiles of temperature and water vapor, and other gas contents. A key issue in the retrieval is the ILL POSED problem (unknowns > equations). Radiative transfer equation L ( ) ( ) B(, T) (1 ( )) R ( ) ag at Unknowns > Equations,an ill-posed problem New method development Because the surface emissivity spectrum is generally smooth enough, it can be represented as a piecewise linear function. The number of unknowns will be efficiently decreased. NEΔT=0.1K NEΔT=0.5K LSE Atmospher ic absorption surface emissivity is smooth piecewise linear function can capture the feature of LSE Errors in temperature (K) 21

22 Current work and achievements in quality assurance group Hyperspectral thermal infrared data processing and retrieval The performances of TES methods for hyperspectral TIR data are also influenced by the radiometric, spectral and other properties of the hyperspectral sensors. The influence is necessary to be analyzed! Five published TES method: Iterative spectrally smooth temperature emissivity separation method (ISSTES) (Borel, 1997) Automatic retrieval of temperature and emissivity using spectral smoothness method (ARTEMISS) (Borel, 2008) Spectral smoothness method (SpSm) (Kanani et. al, 2007) Downwelling radiance residual index method (DRRI) (Wang et. al, 2008) linear spectral emissivity constraint method (LSEC) (Wang et. al, 2011) Test data: Simulated hyperspectral TIR data Test aim: Influence of spectral shifting or FWHM broadening on the TES accuracy. All methods are affected by spectral shifting and FWHM broadening but in different magnitude. LSEC (wang et. al, 2011) was the most insensitive method to the spectral shifting and FWHM broadening of the sensor. So, it may be the best choice among the published method. But, much more analysis are needed! T (T ret -T act )(K) T (T ret -T act )(K) ISSTES ARTEMISS SpSm LSEC DRRI (a) (cm -1 ) ISSTES ARTEMISS SpSm LSEC DRRI 0% 5% 10% 20% (b) FWHM The cyan bar is for LSEC 22

23 Prospect The extension of the Cal&Val test site New equipment developing and distribution A complete target system for performance assessment Supporting the calibration and validation for VIS, NIR, SWIR, TIR and LiDAR payloads Technology and method research Test site basis data collection and database construction A multi level and multi scale calibration and validation system building Accurate vicarious calibration methods research Standard airborne payload Satellite Satellite data Airplane Cal. Airborne data Laboratory calibration system Spectrometer imager Targets in sites Cal. In situ measurement 23

24 WGCV 36 The 36th CEOS Working Group on Calibration and Validation Plenary (WGCV 36) Hosted by the Academy of Opto Electronics (AOE) and, the Shanghai Engineering Centre for Microsatellites (SECM), Chinese Academy of Sciences (CAS) Shanghai, China, May 13 17, 2013 There is 34 registrations to date (not including AOE, SECM, etc.) Conference and hotel Location: The Okura Garden Hotel Shanghai Address: 58 Maoming Road (S.), Shanghai, China 24

25 WGCV 36 If you require transportation from the airport or train station to the hotel when you arrive in Shanghai, please send your travel information by e mail to the WGCV Secretariat (Eric.Arsenault@asc csa.gc.ca) by April 30th or sooner if possible. We will arrange for someone to greet you at the arrival gate with a sign indicating WGCV. We will also be providing transportation from the hotel to the airport or train station for your departure from Shanghai. Please also make sure to tell us if you travelling with family or other individuals to make sure that we have enough room to accommodate everyone in the vehicle. Preliminary plan of social events in the evening banquet by AOE, SECM Shanghai acrobatics boat tour 25

26 26